The treatment of large cartilage lesions is a difficult clinical problem for which there are few good solutions. Left untreated, these lesions tend to degenerate to chronic pain and osteoarthritis (OA), ultimately requiring a total joint replacement. For patients suffering from knee OA, and, in particular, unicompartmental OA, unicondylar knee arthroplasty (UKA) is an available first line treatment option that provides many potential advantages over more common total knee replacement procedures. However, this approach remains controversial due to variable clinical results and high rates of revision associated with current implant designs. To overcome the pitfalls associated with current implant designs, we have developed a novel tissue engineering therapy for unicondylar resurfacing for knee OA. The overall goals of this study are two-fold: (Phase I) to repair large osteochondral defects in the knee, for which current treatment paradigms are currently contraindicated, and (Phase II) to resurface a diseased femoral condyle in an ovine model of unicompartmental OA. Our approach for both phases is based on a high-performance three-dimensionally (3D) woven scaffold that mimics the biomechanical properties of native articular cartilage at the initial time of cell seeding, thus providing a highly functional implant. Its unique woven architecture forms to the anatomical curvature of the condyle, provides an environment that encourages cell growth and differentiation, and is capable of integrating with subchondral bone. In Phase I, efficacy of the technology will be evaluated in a large, full-thickness cartilage defect located in the medial condyle of the knee. The following groups will be evaluated: 1) a defect only control group, 2) an acellular scaffold treatment group, 3) autologous mesenchyme derived stem cell (MSC)-seeded scaffold group, and 4) a tissue-engineered treatment group (ex vivo cultured scaffold with autologous MSCs). At 6 months, the regenerative response of all groups will be evaluated histologically and biomechanically. The most successful treatment will then be translated to Phase II where an anatomically condyle-shaped implant will be used to treat OA. All animals in Phase II will be evaluated at 3, 6, and 12 months following repair. Functional measurements will be taken pre- and postoperatively to evaluate joint function and comfort, while sequential radiographs and MRI will be used to assess any morphological changes. Histological and biomechanical properties of the joint tissues collected from the treated joint will be compared to those from the contralateral limb (negative control) to quantify degradative changes. Serum, synovial fluid, and synovium will be analyzed for biomarkers of OA, as well as for adverse inflammatory reactions and to test for wear debris in the joint. This study will primarily provide valuable data on the ability of out technology to treat a range of cartilage pathology, ranging from large cartilaginous defects to unicompartmental OA. Secondarily, the findings of the study will provide insight into clinical, imaging, and serum/synovial fluid biomarkers that may provide additional information on the predictive validity of such measures in knee OA.